Inside EUV Resists

Andrew Grenville, chief executive of resist maker Inpria, sat down with Semiconductor Engineering to talk about photoresists for extreme ultraviolet (EUV) lithography. What follows are excerpts of that conversation.

SE: Photoresists are a critical part of lithography. Resists are light-sensitive materials. They form patterns on a surface when exposed to light. For EUV, they are critical. What is the main difference between traditional chemically amplified resists (CAR) and metal oxide resists for EUV?

Grenville: Inpria’s metal oxide photoresists are built on three complementary design principles, which taken together provide game-changing performance relative to CAR for EUVL. First, our resists are based on a tin-oxide structure which captures EUV photons ~5X more efficiently than traditional resists, thereby minimizing the variations arising from amplification. Second, given their higher absorption, our materials can be made from 5X smaller molecular building blocks while avoiding the need for the amplification that induces a substantial resist blur and degrades resolution. Finally, by creating the pattern in a metal oxide, the etch resistance is 10-100X that of a conventional resist, enabling substantial process simplification with an improved process window.

SE: Why do we need metal oxide resists for EUV? Is CAR running out of steam?

Grenville: It is generally acknowledged that CAR materials are struggling to achieve the required patterning targets, especially with a manufacturable process window. To avoid pattern collapse, resists must be made thinner, but that leads to insufficient etch resistance and variability due to lack of absorbance. Metal oxide materials provide a solution to these constraints as discussed above, with demonstrated extendibility for future generations of EUVL.

SE: For EUV, the industry has been talking about breaking the so-called RLS trade off. This involves resolution, line-edge roughness and sensitivity. It is hard to get good simultaneous RLS with EUV resists. What about the RLS?

Grenville: In short, Inpria materials have been shown to have substantially larger process windows with improved RLS, and that’s before the huge dividends in etch from a hardmask-like resist. Imec showed some work on this at the recent EUVL Symposium.

SE: Inpria has showed 20mJ/cm2 resists, right? Is that enough to do EUV in production?

Grenville: Yes, we have demonstrated 20mJ/cm2 at 16nm hp. This is sufficient for introduction, especially given the recent improvements in source power demonstrated by ASML. We have a roadmap to continue to improve RLS, and materials in the pipeline to support this roadmap. For an independent imaging reference and comparison with CAR, see slide 21 of this deck, for 16nm hp and 13nm hp.

SE: Is LER a problem with resists? Is there a solution?

Grenville: Yes, LER is a problem with conventional resists. The usual tradeoff involves either a) higher dose, or b) inducing blur to smooth the image at the expense of resolution. Our approach has demonstrated improved LER at high resolution.

SE: Is a 250 Watt power source enough with today’s resists? We may need more dose. So wouldn’t we need a source that goes beyond 250 Watts?

Grenville: 250 Watts appears to be sufficient, but the details depend a lot on the specifics of the layer and patterning requirements.

Related Stories
7nm Lithography Choices
EUV: Cost Killer Or Savior?
Is EUV Making Progress?
Gaps Remain For EUV Masks
Resist Sensitivity, Source Power, And EUV Throughput